We propose to discover novel markers of genetic susceptibility for breast cancer using the combined resources of four large on-going epidemiologic studies of African American women. Our collaboration includes the Carolina Breast Cancer Study, Women's Circle of Health Study, Black Women's Health Study, and the Multiethnic Cohort. DNA samples from a total of 5534 African American cases and 5534 controls will be available for analysis, along with in-person interview data and tumor blocks from cases. Three avenues of investigation are proposed: (1) Dense genotyping and fine-mapping of loci identified in previous genome-wide association studies (GWAS) of breast cancer and related candidate genes, with a particular focus on fine-mapping of loci relevant to African American women. Included in the fine-mapping is a unique locus on chromosome 5q31 we recently identified in a GWAS of African American women;(2) Discovery of potential functional alleles using targeted DNA resequencing and gene expression-based approaches;(3) Statistical analyses aimed at identifying genetic risk factors for subgroups of breast cancer cases defined by early age at onset and tumor biology (ER status and intrinsic subtypes: basal-like, luminal A, luminal B, HER2+/ER-). The combined resources of four epidemiologic studies will facilitate identification of at-risk alleles and haplotypes, permit rapid replication of findings, and produce more precise estimates of effect for breast cancer subgroups. Genotyping and other laboratory studies will be conducted in cooperation with the Biospecimen Core, recruitment and enrollment of additional study participants will be undertaken by the Data Collection Core, and statistical analyses will be conducted in cooperation with the Biostatistics and Data Management Core. Genotype data from Project 1 will be used in conjunction with Projects 2, 3 and 4 to develop complex models for breast cancer susceptibility that incorporate genetic markers, modifiable environmental risk factors, and breast cancer tumor subtypes as distinct disease outcomes.